Traingular and bernoulli uncertainty distributions updates (#1009)

Forces the "average" values in the bernoulli and triangular uncertainty distributions to lie exclusively within the range of the minimum and maximum provided values. Exceptions to this lead to an invalidated finish button, forcing users to select a value within the allowable range.

activity_browser/ui/wizards/uncertainty.py

@@ -2,7 +2,7 @@
 from PySide2 import QtCore, QtGui, QtWidgets
 from PySide2.QtCore import Signal, Slot
 import numpy as np
-from stats_arrays import uncertainty_choices as uc
+from stats_arrays import uncertainty_choices as uncertainty
 from stats_arrays.distributions import *
 
 from ..figures import SimpleDistributionPlot
@@ -171,7 +171,7 @@ def __init__(self, parent=None):
         box1 = QtWidgets.QGroupBox("Select the uncertainty distribution")
         box1.setStyleSheet(style_group_box.border_title)
         self.distribution = QtWidgets.QComboBox(box1)
-        self.distribution.addItems([ud.description for ud in uc.choices])
+        self.distribution.addItems([ud.description for ud in uncertainty.choices])
         self.distribution.currentIndexChanged.connect(self.distribution_selection)
         self.registerField("uncertainty type", self.distribution, "currentIndex")
         self.pedigree = QtWidgets.QPushButton("Use pedigree")
@@ -339,7 +339,7 @@ def distribution_selection(self):
         See https://stats-arrays.readthedocs.io/en/latest/index.html for which
         fields to show and hide.
         """
-        self.dist = uc.id_dict[self.distribution.currentIndex()]
+        self.dist = uncertainty.id_dict[self.distribution.currentIndex()]
 
         # Huge if/elif tree to ensure the correct fields are shown.
         if self.dist.id in {0, 1}:
@@ -359,19 +359,24 @@ def distribution_selection(self):
         self.special_distribution_handling()
         self.generate_plot()
 
-    @property
-    def active_fields(self) -> tuple:
-        """Returns anywhere from 0 to 3 fields"""
+    def completed_active_fields(self) -> bool:
+        """Returns a boolean value based on the distribution id.
+        If the distribution contains an average, minimum and maximum this forces the
+        average to exist exclusively within these bounds"""
+        completed = False
         if self.dist.id in {0, 1}:
-            return ()
+            completed = True
         elif self.dist.id in {2, 3}:
-            return self.loc, self.scale
+            completed = all([field.hasAcceptableInput() and field.text() for field in (self.loc, self.scale)])
         elif self.dist.id in {4, 7}:
-            return self.minimum, self.maximum
+            completed = all([field.hasAcceptableInput() and field.text() for field in (self.minimum, self.maximum)])
         elif self.dist.id in {5, 6}:
-            return self.loc, self.minimum, self.maximum
+            completed = all([field.hasAcceptableInput() and field.text() for field in (self.minimum, self.maximum,
+                    self.loc)]) and (float(self.minimum.text()) < float(self.loc.text()) < float(self.maximum.text()))
         elif self.dist.id in {8, 9, 10, 11, 12}:
-            return self.loc, self.scale, self.shape
+            completed = all([field.hasAcceptableInput() and field.text() for field in (self.scale, self.shape,
+                                                                                               self.loc)])
+        return completed
 
     @Slot(name="locToMean")
     def balance_mean_with_loc(self):
@@ -424,10 +429,7 @@ def generate_plot(self) -> None:
 
         Also tests if all of the visible QLineEdit fields have valid values.
         """
-        self.complete = all(
-            (field.hasAcceptableInput() and field.text())
-            for field in self.active_fields
-        )
+        self.complete = self.completed_active_fields()
         no_dist = self.dist.id in {UndefinedUncertainty.id, NoUncertainty.id}
         if self.complete or no_dist:
             array = self.dist.from_dicts(self.wizard().uncertainty_info)